1/39
Looks like no tags are added yet.
Null Hypothesis
H0
No difference or relationship
Goal is to reject this
Alternative Hypothesis
H1
There is a difference in the relationship
Our prediction (can be positive or negative)
Statistical Significance
Is this result likely to be real, or could it have happened by chance?
A statistically significant results means the data gives enough evidence against the null hypothesis (H0)
Judged with P-Value
P-Value
The probability of obtaining results at least as extreme as ours, assuming H0 is true
Conventional threshold: P = 0.05
Smaller P-Value = stronger evidence against H0
Larger P-Value = weaker evidence against H0
False
Not rejecting H0 means we have proved H0 is true
Categorical Variable
Describes which group or category something belongs to
Continuous Variable
Measures how much of something there is
Values are numbers on a scale and can include decimals
Examples:
– Amount spent on lunch: $12.50
– Time spent shopping: 23.5 minutes
– Product weight: 350.2 grams
Chi-Square Goodness of Fit
used to compare observed counts with expected proportions
Amount of Variables: 1 categorical
Example: Do students prefer three restaurants equally?
Chi-Square Test of Independence
used to examine whether two categorical variables are associated
Amount of Variables: 2 categorical
Example: Is class year associated with having a meal plan?
One-Sample T-Test
Used to compare the mean of one group to a known or target value
IV: Fixed reference value
DV: Continuous
Independent Samples T-Test
Used to compare means of two independent groups
IV: Categorical (2 independent groups)
DV: Continuous
Paired Samples T-Test
Used to compare two related measurements from the same people or matched pairs
IV: Categorical (2 related conditions, either before/after or matched measurements)
DV: Continuous
One-Way ANOVA
Used to compare means across 3+ dependent groups
IV: 1 Categorical (3+ groups)
DV: Continuous
Covariates: None
Two-Way ANOVA
(Factorial)
Used to examine two factors and their interactions
IV(s): 2 categorical
DV: Continuous
Covariates: None
ANCOVA
(Analysis of Covariance)
Used to compare group means while statistically adjusting for one or more continuous variables
IV: 1+ categorical
DV: continuous
Covariates: 1+ continuous variables
Repeated Measures ANOVA
Used to compare means of the same people measured 3+ times
IV: 1 categorical (within-subjects, 3+ conditions/time points)
DV: Continuous
Covariates: None
Mean
Arithmetic Average
Median
Middle Value (ordered data)
Mode
Most frequent value
Variance
Spread in squared units
Standard Deviation
Typical spread around the mean
Covary
When the relationship between 2 variables is consistently related
Presence
Are the variables related? Is there a consistent and systematic
relationship?
Direction
Is the relationship positive or negative? As one variable
increases, does the other also increase or does it decrease?
Strength
How much variance do the variables have in common? Is the
relationship weak, moderate or strong?
Type
What is the nature of the relationship? Does the relationship
change over the range of the variables, or does it stay
constant
Correlation
A standardized covariance
Pearson Correlation Coefficient
(R)
A statistical measure of the strength of a linear relationship between two variables
ranges from -1 to 1
• A t-test statistic is calculated to determine the p-value
• Assumptions
• Data is interval or ratio (e.g. continuous measure 1-7 pt etc…)
• Relationship is linear
• Normally distributed population parameters
Regression
a statistical technique that analyzes the linear relationship between two variables by estimating coefficients for an equation for a straight line
Bivariate Regression
AKA simple linear regression (SLR)
The equation is Y= a + bX + e, where
• Y = the dependent variable (actual values in your dataset)
• a = the intercept (calculated)
• b = the slope (calculated)
• X = the independent variable (actual values in your dataset)
• e = error
Y
The dependent variable (actual values in your dataset)
a
The intercept (calculated)
b
The slope (calculated)
X
The independent variable (data in your data set)
e
error
Large b
strong predictor of y
R²
Variance explained
Simple regression
1 predictor
Multiple Regression
2+ predictors